Description. Section: Surgery Effective Date: January 15, 2015 Subsection: Transplant Original Policy Date: September 13, 2012 Subject:

Transcription

1 Last Review Status/Date: December 2014 Page: 1 of 26 Description Mechanical devices to assist or replace a failing heart have been developed over many decades of research. A ventricular assist device (VAD) is a mechanical support attached to the native heart and vessels to augment cardiac output. The VAD may be used as a bridge to heart transplantation or as destination therapy in those who are not candidates for transplantation. The VAD has also been used as a bridge to recovery in patients with reversible conditions affecting cardiac output. Background Heart failure may be the consequence of a number of differing etiologies, including ischemic heart disease, cardiomyopathy, congenital heart defects, or rejection of a heart transplant. The reduction of cardiac output is considered to be severe when systemic circulation cannot meet the body s needs under minimal exertion. Heart transplantation improves quality of life and has survival rates at 1, 5, and 10 years of 89%, 75%, and 56%, respectively. (1) The supply of donor organs has leveled off, while candidates for transplants are increasing, compelling the development of mechanical devices. Left (LVAD) Implantable ventricular assist devices are attached to the native heart, which may have enough residual activity to withstand a device failure in the short term. In reversible conditions of heart failure, the native heart may regain some function, and weaning and explanting of the mechanical support system after months of use has been described. Ventricular assist devices can be classified as internal or external, electrically or pneumatically powered, and pulsatile or continuous flow. Initial devices were pulsatile, mimicking the action of a beating heart. Devices that are more recent may utilize a pump, which provides continuous flow. Continuous devices may move blood in rotary or axial flow. Surgically implanted ventricular assist devices represent a method of providing mechanical circulatory support for patients not expected to survive until a donor heart becomes available for transplant or for whom transplantation is otherwise contraindicated or unavailable. They are most commonly used to support the left ventricle, but right ventricular and biventricular devices may be

2 Page: 2 of 26 used. The device is larger than most native hearts, and therefore the size of the patient is an important consideration: the pump may be implanted in the thorax or abdomen or remain external to the body. Inflow to the device is attached to the apex of the failed ventricle, while outflow is attached to the corresponding great artery (aorta for left ventricle, pulmonary artery for right ventricle). A small portion of ventricular wall is removed for insertion of the outflow tube; extensive cardiotomy affecting the ventricular wall may preclude VAD use. Devices in which the majority of the system s components are external to the body are for shortterm use (6 hours to 14 days) only, due to the increased risk of infection and need for careful, inhospital monitoring. Some circulatory assist devices are placed percutaneously, i.e., are not implanted. These may be referred to as percutaneous VADs. Percutaneous (pvad) pvads have been developed for short-term use in patients who require acute circulatory support. These devices are placed through the femoral artery. Two different pvads have been developed, the TandemHeart (Cardiac Assist, Pittsburgh, PA), and the Impella device (AbioMed, Aachen, Germany). In the TandemHeart system, a catheter is introduced through the femoral artery and passed into the left atrium via transseptal puncture. Oxygenated blood is then pumped from the left atrium into the arterial system via the femoral artery. The Impella device is also introduced through a femoral artery catheter. In this device, a small pump is contained within the catheter that is placed into the left ventricle. Blood is pumped from the left ventricle, through the device, and into the ascending aorta. Adverse events associated with pvad include access site complications such as bleeding, aneurysms, or leg ischemia. Cardiovascular complications can also occur, such as perforation, myocardial infarction (MI), stroke, and arrhythmias. There are several situations in which pvad may offer possible benefits: 1) cardiogenic shock that is refractory to medications and intra-aortic balloon pump (IABP), 2) cardiogenic shock, as an alternative to IABP, 3) high-risk patients undergoing invasive cardiac procedures who need circulatory support, and 4) patients completing cardiotomy surgery to allow the heart muscle to heal. Regulatory Status Total Artificial Heart In October 2004, device CardioWest Temporary Total Artificial Heart (SynCardia Systems Inc., Tucson, AZ) was approved by the U.S. Food and Drug Administration (FDA) through the premarket approval process (PMA) for use as a bridge to transplant in cardiac transplant-eligible candidates at risk of imminent death from biventricular failure. Also, the temporary CardioWest Total Artificial Heart (TAH-t) is intended for use inside the hospital. In April 2010, FDA approved a name-change to SynCardia Temporary Total Artificial Heart.

3 Page: 3 of 26 In September 2006, device AbioCor Implantable Replacement Heart System (AbioMed Inc., Danvers MA) was approved by FDA through the Humanitarian Device Exemption (HDE) process for use in severe biventricular end-stage heart disease patients who are not cardiac transplant candidates and who: are younger than 75 years of age require multiple inotropic support are not treatable by left ventricular assist device (LVAD) destination therapy; and are not weanable from biventricular support if on such support. In addition to meeting other criteria, patients who are candidates for the AbioCor TAH must undergo a screening process to determine if their chest volume is large enough to hold the device. The device is too large for approximately 90% of women and for many men. FDA is requiring the company to provide a comprehensive patient information package to patients and families. To further refine and improve the use of this artificial heart technology, AbioMed will conduct a postmarketing study of 25 additional patients. The postmarketing study was recommended by the Circulatory Systems Devices Panel, a part of FDA's Medical Devices Advisory Committee. In December 1995, Thoratec Ventricular Assist Device System (Thoratec Corp., Pleasanton, CA) was approved by the FDA through the premarket approval process for use as a bridge to transplantation in patients suffering from end-stage heart failure. The patient should meet all of the following criteria: 1. candidate for cardiac transplantation; 2. imminent risk of dying before donor heart procurement; and 3. dependence on, or incomplete response to, continuous vasopressor support. April 25, 2012 the FDA Advisory Committee voted to recommend device approval as a bridge to transplant for end stage heart failure for the HeartWare (HeartWare Int., Inc.) intrapericardial, continuous-flow, centrifugal pump device. This device remains investigational until FDA approval has been granted. Since the initial device was approved, many others have been approved for marketing or by the human device exemption (See Table below). Percutaneous (circulatory assist devices) The TandemHeart (Cardiac Assist, Pittsburgh) received FDA 510(k) approval in May 2005 for short-term (less than 6 hours) use in patients requiring circulatory support. The Impella Recover LP 2.5 Percutaneous Cardiac Support System (Abiomed, Aachen, Germany) received a similar 510(k) approval for short-term circulatory support in September 2008.

4 Page: 4 of 26 Since the initial device was approved, many others have been approved for marketing or by the human device exemption (See Table below). VAD Device Manufacturer Date of Initial Approval Method of FDA clearance Indication HeartMate XVE Thoratec 2003 HDE Pulsatile Thoratec IVAD DeBakey VAD Child Thoratec August 2004 PMA Supplemen t Bridge to Transplant and postcardiotomy. Authorized only for internal implant, not for paracorporeal implant due to reliability issues. MicroMed April 2004 HDE Bridge to Transplant in children 5 16 years of age HeartMate II Thoratec April 2008 PMA Bridge to Transplant and Destination Centrimag Levitronix October 2008 HVAD HeartWare October 2008 Stratos LV CRT-P and Stratos LV-T CRT-P HDE PMA Post cardiotomy Initiated US BTT trial in October 2008 (completed February 2010) and US DT trial in August Miniature "third generation" device with centrifugal blood path and hydromagnetically suspended rotor that may be placed in the pericardial space. Biotronik, Inc. May 2008 PMA For patients who have moderate to severe heart failure (NYHA class II l/iv), including left ventricular dysfunction (EF<35%) and QRS >120 ms and remain symptomatic. Contak Guidant Corp May 2008 PMA For patients with moderate to severe heart failure who remain symptomatic despite stable, optimal heart failure drug therapy and have left ventricular dysfunction.

5 Page: 5 of 26 HeartMate III Thoratec 2010 PMA Supplemen t EXCOR Berlin Heart July 21, 2011 Thoratec PVAD (Paracorporeal Ventricular Assist Device) HDE Continuous flow driven by a magnetically suspended axial flow rotor. Pediatric VAD Thoratec (k) Pulsatile system includes three major components: Blood pump, cannulae and pneumatic driver. Received FDA approval for BTT in 1995 and for postcardiotomy recovery in TandemHeart Cardiac Assist September 2005 Impella LP 2.5, 5.0, &LD HeartWare Ventricular Assist System 510(k) Temporary left ventricular bypass of 6 hours or less. Abiomed May (k) Partial circulatory support using an extracorporeal bypass control unit for periods up to 6 hours. HeartWare December 2012 PMA Bridge to transplant Source: U. S. Food and Drug Administration, Medical Devices. Several other devices are in clinical trials or awaiting FDA/Congressional review. Related Policies Heart/Lung Transplant Heart Transplant Policy *This policy statement applies to clinical review performed for pre-service (Prior Approval, Precertification, Advanced Benefit Determination, etc.) and/or post-service claims. Post-Cardiotomy Setting/Bridge to Recovery Implantable ventricular assist devices with FDA approval or clearance may be considered medically necessary in the post-cardiotomy setting in patients who are unable to be weaned off cardiopulmonary bypass.

6 Page: 6 of 26 Bridge to Transplantation Implantable ventricular assist devices with FDA approval or clearance may be considered medically necessary as a bridge to heart transplantation for patients who are currently listed as heart transplantation candidates and not expected to survive until a donor heart can be obtained, or are undergoing evaluation to determine candidacy for heart transplantation. Ventricular assist devices with FDA approval or clearance, including humanitarian device exemptions, may be considered medically necessary as a bridge to heart transplantation in children aged 5 to 16 years who are currently listed as heart transplantation candidates and not expected to survive until a donor heart can be obtained, or are undergoing evaluation to determine candidacy for heart transplantation. Total artificial hearts with FDA-approved devices may be considered medically necessary as a bridge to heart transplantation for patients with biventricular failure who have no other reasonable medical or surgical treatment options, who are ineligible for other univentricular or biventricular support devices, and are currently listed as heart transplantation candidates or are undergoing evaluation to determine candidacy for heart transplantation, and not expected to survive until a donor heart can be obtained. Destination Therapy Implantable ventricular assist devices with FDA approval or clearance may be considered medically necessary as destination therapy with end-stage heart failure patients who are ineligible for human heart transplant and who meet the following REMATCH Study criteria: New York Heart Association (NYHA) Class IV heart failure for >60 days, OR patients in NYHA Class III/IV for 28 days, received >14 days support with intra-aortic balloon pump or dependent on IV inotropic agents, with 2 failed weaning attempts. In addition, patients must not be candidates for human heart transplant for 1 or more of the following reasons: Age >65 year; or Insulin-dependent diabetes mellitus with end-organ damage; OR Chronic renal failure (serum creatinine >2.5 mg/dl for >90 days; OR Presence of other clinically significant condition. Other Indications Other applications of implantable ventricular devices or total artificial hearts are considered not medically necessary. The use of non-fda approved or cleared implantable ventricular assist devices is considered investigational.

7 Page: 7 of 26 Percutaneous ventricular assist devices (pvads) are considered not medically necessary for all indications. Policy Guidelines Only two ventricular assist device (VAD) has approval from the U.S. Food and Drug Administration (FDA) for the pediatric population. The DeBakey VAD Child device has FDA approval (HDE process) for use in children ages 5 to 16 years who are awaiting a heart transplant, i.e., as a bridge to transplant. The Berlin EXCOR VAD is indicated for children with severe isolated left ventricular or biventricular dysfunction who are candidates for cardiac transplant and require circulatory support. In general, candidates for bridge-to-transplant implantable ventricular assist devices (VADs) are those who are considered appropriate heart transplant candidates but who are unlikely to survive the waiting period until a human heart donor is available. Some studies have included the following hemodynamic selection criteria: either a left atrial pressure of 20 mm Hg or a cardiac index of <2.0 L/min/m while receiving maximal medical support. Patients with VADs are classified by the United Network for Organ Sharing (UNOS) as Status I, that is, persons who are most ill and are considered the highest priority for transplant. The median duration for time on the device is between 20 and 120 days. Contraindications for bridge to transplant VADs include conditions that would generally exclude patients for heart transplant. Such conditions are chronic irreversible hepatic, renal, or respiratory failure; systemic infection; coagulation disorders, and inadequate psychosocial support. Due to potential problems with adequate function of the VAD, implantation is also contraindicated in patients with uncorrected valvular disease. See policy Heart Transplantation for further discussion of heart transplant candidancy. In addition, patients must have sufficient space in the thorax and/or abdominal cavity for the device. In the case of the CardioWest temporary Total Artificial Heart, this excludes patients with body surface areas less than 1.7 m2 or who have a distance between the sternum and 10th anterior rib of less than 10 cm as measured by computed tomography scan. Rationale Literature Review The literature review focuses on 3 situations: short-term use of the ventricular assist device (VAD) for cardiac shock following cardiac surgery, use of the VAD as a bridge to heart transplant, and use of the VAD as destination therapy in end-stage heart failure.

8 Bridge to Recovery: Post-cardiotomy Setting Page: 8 of 26 Five studies of the Centrimag Right Ventricular Assist Device (RVAS) included between 12 and 32 patients, the majority of whom received biventricular devices. (2-4) Indications (and numbers of patients) in these 5 studies were: support for post-cardiotomy cardiogenic shock (bridge to recovery, n=53), bridge to long-term device implantation (n=9), treatment of right heart failure in patients who previously received left ventricular assist devices (LVADs) (n=15), bridge to later decision when neurologic status is clarified (n=16), and acute donor graft failure (n=6). The mean time on mechanical circulatory support ranged from 9.4 days to 46.9 days. The 30-day mortality rates were between 17% and 63%. The proportion of patients discharged from the hospital was between 30% and 83%. Major complications included bleeding requiring reoperation, sepsis, and stroke. No device failures were observed in these studies. LVADs may have a role in bridging patients to recovery, particularly if there is reverse remodeling of the left ventricle. A number of relatively small, noncomparative studies have evaluated LVADs as bridge-to- recovery therapy. In a 2006 study, a series of 15 patients with severe heart failure due to nonischemic cardiomyopathy underwent implantation of LVADs, along with medical management designed to enhance myocardial recovery. Eleven of 15 patients had enough myocardial recovery to undergo LVAD explantation; 2 patients died after explantation. Among those who survived, the cumulate rate of freedom from recurring heart failure was 100% and 88.9%, respectively, at 1 and 4 years postexplantation. The same group subsequently reported results of their LVAD explantation protocol among patients with severe heart failure due to nonischemic cardiopathy who had nonpulsatile LVADs implanted.(5) They included 20 patients who received a combination of angiotensin converting enzyme ACE-inhibitors, beta-blockers, and adosterol antagonists followed by the β2-agonist clenbuterol. One patient was lost to follow-up and died after 240 days of support. Of the remaining 19 patients, 12 (63.2%) were successfully explanted after a mean 286 days; estimated survival without heart failure recurrence was 83.3% at 1 and 3 years. In a prospective multicenter study to assess myocardial recovery in patients with LVAD implantation as a bridge to transplant, Maybaum et al evaluated 67 patients with heart failure who had undergone LVAD implantation for severe heart failure.(6) After 30 days, patients demonstrated significant improvements compared with pre-lvad state in left ventricular ejection fraction (17.1% vs 34.12%, p<0.001), left ventricular end-diastolic diameter (7.1 cm vs 5.1 cm, p<0.001), and left ventricular mass (320 g vs 194 g, p<0.001). However, only 9% of patients demonstrated enough recovery to have their LVAD explanted. Section Summary: The studies previously outlined indicate that a subset of patients who receive an LVAD as a bridge to transplant or as destination therapy demonstrate improvements in their cardiac function, sometimes to the point that they no longer require the LVAD. However, questions remain about defining and identifying the population most likely to experience cardiac recovery with LVAD placement. One clearly-defined population in which the potential for myocardial recovery exists is in the postcardiotomy setting. The current evidence is insufficient to allow the identification of other heart failure patient populations who might benefit from the use of an LVAD as a specific bridge-to-recovery treatment strategy. Ongoing research studies are addressing this question, along with protocols for transitioning patients off LVAD use.

9 Bridge to Transplant: Left Page: 9 of 26 A 1996 Blue Cross and Blue Shield Association Technology Evaluation Center (TEC) assessment concluded that left ventricular assist devices (LVADs) could provide an effective bridge to transplantation. (7) Goldstein and colleagues published a more recent review. (8) It should be recognized that LVADs do not change the number of patients undergoing heart transplantation due to the fixed number of donor hearts. However, the VAD will categorize its recipient as a highpriority heart transplant candidate. Published studies continue to report that the use of a VAD does not compromise the success of a subsequent heart transplant and, in fact, may improve post-transplant survival, thus improving the use of donor hearts. (9-12) Currently available implantable LVADs consist of pulsatile devices that require stiff power vent lines that perforate the skin and implantable pump chambers, as well as non-pulsatile axial flow systems of smaller size and lower noise levels. (13) In 5 reports, with samples ranging from 32 to 279 patients, most participants received the continuous-flow device as a bridge to transplantation. (14-18) Survival rates at 6 months were between 67% and 87%, and between 50% and 80% at 1 year. These rates are similar to those observed in a recent report of a federal circulatory support device registry. (19) An additional report from INTERMACS comparing the HeartMate II with other LVAD devices for patients who received them with a bridge to transplantation indication reported that 91% and 80% of HeartMate II and other LVAD patients, respectively, reached transplant, cardiac recovery, or ongoing LVAD support by 6 months.(20). A study by Patel and colleagues compared HeartMate I and HeartMate II recipients at a single center, finding the same 1-year survival and similar rates of subsequent development of right heart failure. (17) Serious adverse events occurring after HeartMate II-implantation include bleeding episodes requiring reoperation, stroke, infection, and device failure. A systematic review published in 2011 examined the evidence on the effect of LVADs on posttransplant outcomes. (21) This review included 31 observational studies that compared outcomes of transplant in patients who did and did not have pretransplant LVAD. Survival at 1 year was more likely in patients who had LVAD treatment, but this benefit was confined to patients who received an intracorporeal device (relative risk [RR]=1.8, 95% confidence interval [CI], 1.53 to 2.13). For patients treated with an extracorporeal device, the likelihood of survival was not different from patients who were not treated with an LVAD (RR=1.08, 95% CI, 0.95 to 1.22). There was no difference in the risk of rejection between patients who did and did not receive LVAD treatment. Additional studies published since the systematic review continue to report high success rates in bridging to transplant. In 2011, Strueber et al (22) published a case series of 50 patients awaiting heart transplantation treated with HeartWare VAD. Patients were followed until transplantation, myocardial recovery, device explant, or death. The median duration of time on the LVAD was 322 days. Nine patients died; 3 from sepsis, 3 from multiple organ failure, and 3 from hemorrhagic stroke. At the end of follow-up, 20 patients had undergone transplant (40%), 4 had the pump explanted (8%), and the remaining 17 continued on pump support (34%). The most common

10 Page: 10 of 26 complications were infection and bleeding. A total of 21 patients had infections (42%), and 5 patients had sepsis (10%). Bleeding complications occurred in 15 patients (30%), 10 of whom (20%) required surgery for bleeding. In 2012, Aaronson et al reported results of a multicenter, prospective study of a newer generation LVAD, the HeartWare, which is a smaller, continuous flow centrifugal device that is implanted in the pericardial space.(23) The study enrolled 140 patients who were awaiting heart transplantation who underwent HeartWare implantation. A control group of 499 subjects comprised patients drawn from the INTERMACS database, which collects data on patients who receive FDA-approved durable mechanical circulatory support devices. The study s primary outcome was defined as survival on the originally implanted device, transplantation, or explantation for ventricular recovery at 180 days. Secondary outcomes were comparisons of survival between groups and functional, quality of life, and adverse event outcomes in the HeartWare group. Success occurred in 90.7% of the HeartWare group and 90.1% of controls (p<0.001, noninferiority with a 15% margin). Serious adverse events in the HeartWare group included, most commonly, bleeding, infections, and perioperative right heart failure. In 2013, Slaughter et al reported combined outcomes for patients included in the HeartWare bridge-to-transplant study previously described and a continued-access protocol granted by FDA.(24) The study included 322 patients with heart failure, eligible for heart transplant, who received the HeartWare (140patients from the original study; 190 patients in the continueaccess protocol who were monitored to outcome or had completed 180 days of follow-up at the time of this analysis. Survival at 60, 180, and 360 days was 97%, 91%, and 84%, respectively. The most common adverse events were respiratory dysfunction, arrhythmias, sepsis, and driveline exit site infections. Patients generally had improvements in quality-of-life measures. Pediatric patients. There is one FDA-approved device, the EXCOR Pediatric VAD, via the HDE process, available for use as a bridge to cardiac transplant in children. This HDE approval was based on data from children who were a part of the initial clinical studies of this device. (25) Publications have reported positive outcomes for children using VADs as a bridge to transplantation. Using the UNOS database, Davies et al reported on use of VADs in pediatric patients undergoing heart transplantation. (26) Their analysis concluded that pediatric patients requiring a pretransplantation VAD have similar long-term survival to those not receiving mechanical circulatory support. Following the FDA approval, Fraser et al evaluated the EXCOR device among 48 children, aged 16 or younger with 2-ventricle circulation, who had severe heart failure, despite optimized treatment and were listed for heart transplant.(27) Patients were divided into 2 groups based on body surface area. A historic control group of children receiving circulatory support with extracorporeal membrane oxygenation (ECMO) from the Extracorporeal Life Support Organization registry, matched in a 2:1 fashion with study participants based on propensity-score matching. For participants in cohort 1 (body surface area <0.7 m2), the median survival time had not been reached at 174 days, while in the matched ECMO comparison group, the median survival was 13 days (p<0.001). For participants in cohort 2 (body surface area 0.7 to <1.5 m2), the median

11 Page: 11 of 26 survival was 144 days, compared with 10 days in the matched ECMO group (p<0.001).rates of adverse events were high in both EXCOR device cohorts, including major bleeding (in 42% and 50% of cohort 1 and cohort 2, respectively), infection (in 63% and 50% of cohort 1 and cohort 2, respectively), and stroke (in 29% of both cohorts). In 2013, Almond et al reported results from a prospective, multicenter registry to evaluate outcomes in children who received the Berlin Heart EXCOR device as a bridge to transplant.(28) This study included a broader patient population than the Fraser et al study. All patients were followed up from the time of EXCOR implantation until transplantation, death, or recovery. The study included 204 children, 67% of whom received the device under compassionate use. Survival at 12 months on EXCOR support was 75%, including 64% who survived to transplantation, 6% who recovered (device explanted and patient survived 30 days), and 5% alive with the device in place. Section Summary. In adults, the evidence on the efficacy of LVADs as bridge to transplant consists of numerous uncontrolled trials and trials comparing different LVAD devices of patients who have no other treatment options. In children, the evidence consists of several uncontrolled trials and 1 trial with historical controls. These studies report that substantial numbers of patients survive to transplant in situations in which survival would not be otherwise expected. Despite the lack of highquality controlled trials, this evidence is sufficient to determine that outcomes are improved in patients who have no other options for survival. The impact of pretransplant LVADs on survival from transplant is uncertain, with some studies reporting worse survival in patients receiving LVADs, but other studies reporting similar or improved survival. LVADS as Destination Therapy LVADs as destination therapy based on the 2002 Blue Cross and Blue Shield Association TEC assessment (29) offers the following observations and conclusions: The available evidence comes from a single, well-designed and rigorously conducted randomized trial, known as the REMATCH study. (30) The study was a cooperative effort of Thoratec, Columbia University, and the National Institutes of Health. The randomized trial found that patients with end-stage heart failure who are not candidates for cardiac transplantation have significantly better survival on a VAD compared with treatment by optimal medical therapy. Median survival was improved by approximately 8.5 months. Serious adverse events were more common in the VAD group, but these appear to be outweighed by this group s better outcomes on function; New York Heart Association (NYHA) class was significantly improved, as was quality of life among those living to 12 months. VAD patients spend a greater relative proportion of time inside the hospital than medical management patients do, but the survival advantage would mean a longer absolute time outside the hospital.

12 Page: 12 of 26 Park et al published an extended 2-year follow-up of patients in the REMATCH trial, which found that survival and quality-of-life benefits were still apparent. In addition, this study and other case series suggest continuing improvement in outcomes related to ongoing improvements in the device and in patient management. (31, 32) However, the durability of the HeartMate device used in the REMATCH trial is a concern; for example, at one participating institution, all 6 long-term survivors required device change-outs. Next generation devices consisting of smaller continuous flow devices are eagerly anticipated. After publication of the REMATCH study results, Rogers et al published results from a prospective, nonrandomized clinical trial comparing LVAD as destination therapy with optimal medical therapy for patients with heart failure who were not candidates for heart transplant.(33) Fifty-five patients who had NYHA functional Class IV symptoms and who failed weaning from inotropic support were offered a Novacor LVAD; 18 of these did not receive a device due to preference or device unavailability and acted as a control group. The LVAD-treated patients had superior survival rates at 6 months (46% vs 22%; p=0.03) and 12 months (27% vs 11%; p=0.02), along with fewer adverse events. Section Summary: The main piece of evidence on the efficacy of LVADs as destination therapy in patients who are not transplant candidates is from a multicenter randomized controlled trial (RCT), the REMATCH study. This trial reported that the use of LVADs led to improvements in survival, quality of life, and functional status. This evidence is sufficient to establish that health outcomes are improved for this patient population. Comparative Efficacy of Continuous Flow versus Pulsatile Flow Devices In December 2009, Slaughter et al published data from an unblinded randomized multicenter trial comparing a continuous flow device with a pulsatile device. (34) Subjects were randomly assigned to continuous-flow or pulsatile-flow devices on a 2:1 block-randomization basis. The primary outcome measured was a composite end point of 2-year survival, free of disabling stroke or need for device replacement. Continuous-flow patients (n=134) reached the primary outcome at a rate of 46% (95% CI, 38 to 55) compared with pulsatile-flow patients (n=66) rate of 11% (95% CI, 3 to 18), which was a significant difference (p<0.001). Analysis of constituent factors indicated that a lower rate of devices needing replacement in the continuous-flow group had the largest effect on the composite end point; 2- year death rate also favored this device (58% vs 24%, respectively; p=0.008). Stroke and death (within 2 years of implantation) were similar in the 2 groups (stroke rate 12% and death rate 36%). Quality-of-life scores were also similar in the 2 groups. Although unblinded, this randomized trial adds to the evidence favoring continuous-flow devices. Nativi et al (35) published a nonrandomized comparison of pulsatile versus continuous flow devices using data from the registry of the International Society for Heart and Lung Transplantation on 8,557 patients undergoing transplant. Comparisons were made among patients receiving a pulsatile LVAD, a continuous flow LVAD, and no LVAD. Two time periods were used for analysis; the first was pre-2004, when nearly all LVADs were pulsatile devices, and post-2004 when continuous use devices began to be used in clinical care. Comparing the first time period to the

13 Page: 13 of 26 second time period, there was a significantly greater risk of mortality in the first time period compared with the second time period RR=1.30, 95% CI, 1.03 to 1.65, p=0.03). When analysis was confined to the second time period, there was no significant improvement in survival for the continuous group compared with the pulsatile group (RR=1.25, 95% CI, 1.03 to 1.65, p=0.03). Other nonrandomized studies that have compared outcomes from different types of LVADs have been smaller and/or focused on physiologic outcomes. (36-39) In some of these studies, the continuous flow devices exhibit greater improvement in physiologic measures, but none of these studies have reported significant differences between devices in clinical outcomes. Section Summary: The evidence on the comparative efficacy of different devices consists of one RCT and several nonrandomized comparative studies. The RCT reported fairly large differences in a composite outcome measure favoring the continuous flow devices, with increases in revision and reoperation rates for the pulsatile device group being the largest factor driving the difference in outcomes. Other nonrandomized comparative studies, including one database study with large numbers of patients, have not reported important differences between devices on clinical outcomes. Total Artificial Hearts TAH as Bridge to Transplant The FDA approval of the CardioWest TAH was based on the results of a nonrandomized, prospective study of 81 patients.(40) Patients had failed inotropic therapy and had biventricular failure and thus were not considered appropriate candidates for an LVAD. The rate of survival to transplant was 79%, which was considered comparable with the experience with LVAD in patients with left ventricular failure. The mean time from entry into the study until transplantation or death was 79.1 days.other case series have been reported on outcomes of the TAH as a bridge to transplant. For example, Copeland et al(41) reported on 101 patients treated with the SynCardia artificial heart as a bridge to transplant. All patients either met established criteria for mechanically assisted circulatory support, or were failing medical therapy on multiple inotropic drugs. The mean support time was 87 days, with a range of 1 to 441 days. Survival to transplant was 68.3% (69/101). Of the 32 deaths before transplant, 13 were due to multiple organ failure, 6 were due to pulmonary failure, and 4 were due to neurologic injury. Survival after transplant at 1, 5, and 10 years, respectively, was 76.8%, 60.5%, and 41.2%. TAH as Destination Therapy Data concerning the artificial heart are available from information concerning the FDA approval(42) and from a published article describing results for the first 7 patients.(43) FDA indicated that their decision was based on the company's laboratory and animal testing and on a small clinical study of 14 patients that was conducted by Abiomed. The patients had a 1-month survival prognosis of not more than 30%, were not eligible for cardiac transplants, and were felt to not benefit from VAD therapy. The study was reported to show that the device is safe and has likely benefit for people

14 Page: 14 of 26 with severe heart failure whose death is imminent and for whom no alternative treatments are available. Of the 14 patients in the study, 12 survived surgery. Mean duration of support for the patients was 5.3 months. In some cases, the device extended survival by several months; survival was 17 months in 1 patient. Six patients were ambulatory; 1 patient was discharged home. Complications included postoperative bleeding and neurologic events. Device-related infection was "non-existent." This device shows technologic progress, and these initial results are encouraging; however, a number of questions remain. These questions may be answered once the results of the 14-patient study are published, or data on a larger group of patients may be needed. One issue is to further analyze relevant patient outcomes (eg, complications, quality of life, and survival). Therefore, based on current information, this device is considered investigational. Section Summary There is a smaller amount of evidence on the use of TAH as a bridge to transplantation, or as destination therapy, compared with the use of LVADs. The type of evidence on bridge to transplant is similar to that for LVADs, ie, case series reporting substantial survival rates in patients without other alternatives. Therefore, this evidence is sufficient to conclude that TAH improves outcomes for these patients similar to LVADs, and is a reasonable alternative for patients who require bridge to transplantation but who are ineligible for other types of support devices. There Is insufficient evidence on the use of TAH as destination therapy to support conclusions. pvads pvads as an Alternative to IABP in Cardiogenic Shock Three RCTs have been published that compare pvads with IABPs for patients with cardiogenic shock, (44-46) along with a systematic review and meta-analysis of these 3 trials. (43) The metaanalysis was published in 2009 by Cheng et al. The 3 RCTs enrolled a total of 100 patients, 53 treated with a pvad and 47 treated with an IABP. All 3 study populations included patients with acute MI and cardiovascular shock; one of the trials (32) restricted this population to patients who were postrevascularization in the acute MI setting. The primary outcomes reported were 30-day mortality, hemodynamic measures of left ventricular (LV) pump function, and adverse events. None of the 3 trials reported an improvement in mortality associated with pvad use. The combined analysis estimated the relative risk for death in pvad patients as 1.06 (95% CI: , p=0.80). All 3 trials reported an improvement in LV hemodynamics in the pvad group. On combined analysis, there was a mean increase in cardiac index of 0.35 L/min/m 2 for the pvad group, an increase in mean arterial pressure of 12.8 mm Hg (95% CI: , p<0.001), and a decrease in pulmonary capillary wedge pressure of 5.3 mm Hg (95% CI: , p<0.05). Complications were more common in the pvad group. On combined analysis, patients in the pvad group had a significantly increased likelihood of bleeding events with a relative risk of 2.35

15 Page: 15 of 26 (95% CI: ). Leg ischemia was also more common in the pvad group, but this difference did not meet statistical significance (relative risk [RR]: 2.59, 95% CI: , p=0.13). Case series of patients treated with pvads as an alternative to IABP in cardiogenic shock have been published, (48) and report high success rates as a bridge to alternative therapies. However, these studies do not add much to the evidence on efficacy that is reported from the RCTs. pvads as Bridge to Recovery in Cardiogenic Shock Refractory to IABP Case series of patients with cardiogenic shock refractory to IABP who were treated with pvad have also been published. In the largest series, Kar et al. (49) treated 117 patients who had severe, refractory cardiogenic shock with the TandemHeart System. Eighty patients had ischemic cardiomyopathy and 37 had nonischemic cardiomyopathy. There were significant improvements in all hemodynamic measures following LVAD placement. For example, cardiac index increased from L/min/m 2 to L/min/m 2 (p<0.001), and the systolic blood pressure (BP) increased from mm Hg to mm Hg (p<0.001). Complications were common post-lvad implantation. Thirty-four patients had bleeding around the cannula site (29.1%) and 35 developed sepsis during the hospitalization (29.9%). Groin hematoma occurred in 6 patients (5.1%); limb ischemia in 4 patients (3.4%); femoral artery dissection or perforation in 2 patients (1.7%); stroke in 8 patients (6.8%); coagulopathy in 13 patients (11.0%). pvads Ancillary Support in High risk Patients Undergoing Invasive Cardiovascular Procedures The PROTECT trial intended to evaluate whether the Impella 2.5 system improved outcomes for patients undergoing high-risk percutaneous coronary intervention (PCI) procedures. PROTECT I (50) was a feasibility study of 20 patients who had left main disease or last patent coronary conduit that required revascularization but who were not candidates for coronary artery bypass graft (CABG) surgery. High-risk PCI was performed using the Impella system for circulatory support. All of the procedures were completed successfully without any hemodynamic compromise during the procedures. There were 2 patient deaths within 30 days (10%) and 2 patients had a periprocedural MI (10%). An additional 2 patients had evidence of hemolysis, which was transient and resolved without sequelae. The PROTECT II trial was planned as an RCT to compare the Impella system with IABP in patients undergoing high-risk PCI procedures. Enrollment was planned for 654 patients from 50 clinical centers. The primary end point was the composite of 10 different complications occurring within 30 days of the procedure, with the authors hypothesizing a 10% absolute decrease in the complication rate for patients in the pvad group. The trial was discontinued prematurely in late 2010 due to futility, after an interim analysis of the first 327 patients enrolled revealed that the primary end point could not be reached. At the point that the data safety and monitoring board stopped the study, 452 patients had been enrolled, and of whom 3 withdrew consent and 1 died. Results were published by O Neill et al in 2012.(51) The study s primary analysis was intention to treat and included all 448 patients randomly assigned to the Impella system (n=225) or IABP (n=223). The primary composite end point of major adverse effects at 30 days occurred in 35.1%

16 Page: 16 of 26 of Impella patients and in 40.1% of the IABP patients (p=0.277). There was no significant difference in the occurrence of in-hospital death, stroke, or MI between the Impella patients and the IABP patients. A few other case series have described pvad use in high-risk patients undergoing an invasive cardiac procedure. Sjauw et al (52) performed a retrospective analysis of 144 consecutive patients undergoing high-risk PCI with pvad support (Impella system) from a European registry. Endpoints included successful device function and incidence of adverse events at 30 days. The device was successfully implanted in all 144 patients. There was one periprocedural death and 8 deaths at 30 days for a mortality rate of 5.5%. Bleeding requiring transfusion or surgery occurred in 6.2% of patients, and vascular access site complications occurred in 4.0%. There was 1 stroke (0.7%) and no MIs were reported. Maini et al (53) performed a similar retrospective analysis of 175 patients undergoing high-risk PCI with pvad support with the Impella 2.5 circulatory support system. The primary safety end point was the incidence of major adverse cardiac events at 30 days. Secondary end points included device safety and efficacy and patient outcomes at 30 days and 12 months. Angiographic revascularization was successful in 99% of patients. At 30-day follow up, the major adverse cardiac event rate was 8%; survival was 96%, 91%, and 88% at 30 days, 6 months, and 12 months, respectively. Secondary safety end points occurring most frequently included acute renal dysfunction (2.8%), hypotension on support (3.4%), ventricular tachycardia, or cardiopulmonary resuscitation (2.8%); other vascular complications such as vessel dissection and arteriovenous fistula (3.4%), hematomas ipsi- or contra-lateral to the device insertion site (8.6%), infection (5.1%), and blood transfusion (9.7%). Section Summary: pvads have been tested in RCTs and uncontrolled studies of patients with cardiogenic shock and in patients undergoing high-risk cardiac interventions. The RCTs do not report a benefit for use of pvads. In addition, both the RCTs and case series report high rates of adverse events that may outweigh any potential benefits. As a result, the evidence on pvads does not support an improvement in health outcomes for patients with cardiogenic shock or in patients undergoing high-risk cardiac interventions. Ongoing trials A search of the website clinicaltrials.gov identified the following ongoing trials that will include outcomes associated with mechanical circulatory support devices: The Remission from Stage D Heart Failure (RESTAGE-HF) trial (NCT ). This is a noncomparative interventional study to determine the proportion of subjects who have enough improvement in ventricular function after undergoing a standardized LVAD plus pharmacologic recovery treatment and testing protocol to allow removal of the LVAD. Eligible subjects will have NYHA Class IIIB/IV heart failure and be eligible for the HeartMate II implant as bridge to therapy or destination therapy. The HeartMate II device will be combined with a standardized protocol of medical therapy to enhance cardiac remodeling, including carvedilol, spironolactone, losartan, and digoxin. Enrollment is planned for 40 patients. The estimated study completion date is December 2017.

17 Page: 17 of 26 A Clinical Trial to Evaluate the HeartWare Ventricular Assist System (ENDURANCE SUPPLEMENTAL TRIAL) (NCT ). This is a prospective, randomized, unblinded controlled trial to assess the incidence of stroke in patients with advanced heart failure receiving the HeartWare ventricular assist system who received optimal blood pressure management and to compare stroke-free success among those receiving the HeartWare device to those receiving other FDA-approved LVADs for destination therapy. Patients will be randomized to either an experimental group and receive the HeartWare device, or an active comparator group. Enrollment is planned for 429. The estimated study completion date is October A Prospective Study to Evaluate the Safety and Efficacy of the EVAHEART LVAS for Use as a Bridge-to-Transplant (NCT ). This is a prospective, noncomparative study to evaluate the safety and efficacy of the EVAHEART left ventricular assist system in patients with end-stage heart failure who are candidates for heart transplant. Enrollment is planned for 20 patients. The estimated study completion date is July Evaluation of the Jarvik 2000 Left Ventricular Assist System with Post-Auricular Connector--Destination Therapy Study (NCT ). This is a prospective, randomized, unblinded controlled trial to compare the Jarvik 2000 Left Ventricular Assist system with Post- Auricular Connector to an active control group treated with the Thoratec HeartMate II for destination therapy in patients with late-stage heart failure who are ineligible for heart transplant. Enrollment is planned for 350 patients. The estimated study completion date is December Practice Guidelines and Position Statements The American College of Cardiology/American Heart Association (ACC/AHA) released guidelines for the management of heart failure in October 2013 that include recommendations related to the use of for mechanical circulatory support, including both durable and nondurable mechanical circulatory support (MCS) devices.(54) The guidelines categorize percutaneous and extracorporeal VADs as nondurable MCS devices. The following Class IIA guidelines are made related to MCS devices: MCS is beneficial in carefully selected patients with stage D heart failure with reduced ejection fraction (HFrEF) in whom definitive management (e.g., cardiac transplantation) or cardiac recovery is anticipated or planned. (Level of Evidence: B) Nondurable MCS, including the use of percutaneous and extracorporeal VADs, is reasonable as a bridge to recovery or bridge to decision for carefully selected patients with HFrEF with acute, profound hemodynamic compromise. (Level of Evidence; B) Durable MCS is reasonable to prolong survival for carefully selected patients with stage D HFrEF. (Level of Evidence: B) The AHA/ACC guidelines note: Although optimal patient selection for MCS remains an active area of investigation, general indications for referral for MCS therapy include patients with LVEF <25% and NYHA class III IV functional status despite GDMT, including, when indicated, CRT, with either

18 Page: 18 of 26 high predicted 1- to 2-year mortality (e.g., as suggested by markedly reduced peak oxygen consumption and clinical prognostic scores) or dependence on continuous parenteral inotropic support. Patient selection requires a multidisciplinary team of experienced advanced HF and transplantation cardiologists, cardiothoracic surgeons, nurses, and ideally, social workers and palliative care clinicians. In 2012, the American Heart Association published recommendations for the use of MCS. (55) These guidelines define nondurable MCS as intraballoon pumps, extracorporeal membrane oxygenation, extracorporeal VADs, and pvads. The following recommendations were made regarding indications for use of MCS, including durable and nondurable devices: MCS for bridge-to-transplant indication should be considered for transplant-eligible patients with end-stage heart failure who are failing optimal medical, surgical, and/or device therapies and at high risk of dying before receiving a heart transplantation. (Class I; Level of Evidence B). Implantation of MCS in patients before the development of advanced heart failure is associated with better outcomes. Therefore, early referral of heart failure patients is reasonable. (Class IIa; Level of Evidence B). MCS with a durable, implantable device for permanent therapy or destination therapy is beneficial for patients with advanced heart failure, high 1-year mortality resulting from heart failure, and the absence of other life-limiting organ dysfunction; who are failing medical, surgical, and/or device therapies; and who are ineligible for heart transplantation. (Class I; Level of Evidence B). Elective rather than urgent implantation of destination therapy can be beneficial when performed after optimization of medical therapy in advanced heart failure patients who are failing medical, surgical, and/or device therapies. (Class IIa; Level of Evidence C). o Urgent nondurable MCS is reasonable in hemodynamically compromised heart failure patients with end-organ dysfunction and/or relative contraindications to heart transplantation/durable MCS that are expected to improve with time and restoration of an improved hemodynamic profile. (Class IIa; Level of Evidence C). o These patients should be referred to a center with expertise in the management of durable MCS and patients with advanced heart failure. (Class I; Level of Evidence C). Patients who are ineligible for heart transplantation because of pulmonary hypertension related to heart failure alone should be considered for bridge to potential transplant eligibility with durable, long-term MCS. (Class IIa; Level of Evidence B). The Heart Failure Society of America published guidelines in 2010 on surgical approaches to the treatment of heart failure. (56) The following recommendations were made regarding left ventricular assist devices: Patients awaiting heart transplantation who have become refractory to all means of medical circulatory support should be considered for a mechanical support device as a bridge to transplant. (Strength of Evidence = B)